Nvidia's GTC 2025 conference in San Jose, California, drew a record-breaking 25,000 attendees, with many workshops, talks, and panels overflowing with people. Amidst the excitement, Nvidia CEO Jensen Huang attempted to project confidence, unveiling powerful new chips, personal "supercomputers," and even cute robots. However, the company's financials and future prospects remain uncertain, with investors reeling from Nvidia's nosediving stock.

Huang's keynote address focused on assuring attendees that demand for Nvidia's chips won't slow down anytime soon. He claimed that nearly the "entire world got it wrong" on traditional AI scaling falling out of vogue, and that power-hungry reasoning models will drive more demand for the company's chips. To support this claim, Huang showcased Nvidia's next line of Vera Rubin GPUs, which he claimed will perform inference at roughly double the rate of Nvidia's current best Blackwell chip.

However, Huang spent less time addressing the threat to Nvidia's business from upstarts like Cerebras, Groq, and other low-cost inference hardware and cloud providers. Nearly every hyperscaler is developing a custom chip for inference, if not training, as well. AWS has Graviton and Inferentia, Google has TPUs, and Microsoft has Cobalt 100. Additionally, tech giants like OpenAI and Meta, which are currently extremely reliant on Nvidia chips, are looking to reduce those ties via in-house hardware efforts.

If these rivals are successful, it will likely weaken Nvidia's stranglehold on the AI chips market. This may have contributed to Nvidia's share price dipping around 4% following Huang's keynote, as investors might have been holding out hope for more significant announcements or an accelerated launch window.

Tariff tensions also loomed large over the conference. Huang claimed that tariffs wouldn't do "significant damage" in the short run, but stopped short of promising that Nvidia would be shielded from the long-term economic impacts. In response, Nvidia has pledged to spend hundreds of billions of dollars on manufacturing in the U.S., which would help the company diversify its supply chains but also represents a massive cost.

In an effort to seed and grow businesses other than its core chips line, Nvidia drew attention to its new investments in quantum computing. The company announced that it would open a new center in Boston, NVAQC, to advance quantum computing in collaboration with leading hardware and software markers. The center will be equipped with Nvidia chips, which the company says will enable researchers to simulate quantum systems and the models necessary for quantum error correction.

Nvidia also sees what it's calling "personal AI supercomputers" as a potential new revenue-maker. The company launched DGX Spark and DGX Station, both of which are designed to allow users to prototype, fine-tune, and run AI models in a range of sizes at the edge. While neither is exactly inexpensive, with prices in the thousands of dollars, Huang boldly proclaimed that they represent the future of the personal PC.

Ultimately, the success of Nvidia's new initiatives and products will depend on customer adoption and the company's ability to navigate the uncertain landscape of AI chip development. As the industry continues to evolve, one thing is clear: Nvidia's dominance is no longer a foregone conclusion.